Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
Update laya-ts/tests/structured.test.ts
Ang Zhang15bbb1507173
diff --git a/laya-ts/tests/structured.test.ts b/laya-ts/tests/structured.test.tsindex 1436f98a..ed415a6e 100644--- a/laya-ts/tests/structured.test.ts+++ b/laya-ts/tests/structured.test.ts@@ -471,6 +471,22 @@ describe("structured/decideBatch", () => {expect((results[0].answers.urgency as any).low_confidence).toBe(true);});+ it("accepts python parity alias min_confidence in decideBatch", async () => {+ const answersWithLowConf = {+ department: { type: "choice", choice: "billing", confidence: 0.9, answer_confidence: 0.95 },+ urgency: { type: "score", score: 2.0, confidence: 0.4, answer_confidence: 0.4 },+ };+ const runner = fakeBatchRunner([answersWithLowConf]);+ const results = await decideBatch(runner, ["s1"], SCHEMA, {+ min_confidence: 0.85,+ returnDetails: true,+ });++ expect(results[0].values.department).toBe("billing");+ expect(results[0].values.urgency).toBeNull();+ expect((results[0].answers.urgency as any).low_confidence).toBe(true);+ });+it("handles empty states array cleanly", async () => {const runner = fakeBatchRunner([ANSWERS]);const results = await decideBatch(runner, [], SCHEMA);